#!python
#
# Remove remaining tags from wiki text.

import os
import glob
import argparse
from wikivector import wiki


def main(input_dir, output_dir):
    os.makedirs(output_dir, exist_ok=True)
    input_files = glob.glob(os.path.join(input_dir, '*.txt'))
    for input_file in input_files:
        filename = os.path.basename(input_file)
        output_file = os.path.join(output_dir, filename)

        # run twice to remove nested tags
        wiki.remove_tags(input_file, output_file)
        wiki.remove_tags(output_file, output_file)


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('input_dir', help="Path to directory with text files")
    parser.add_argument('output_dir', help="Path to directory to write new text files")
    args = parser.parse_args()
    main(args.input_dir, args.output_dir)
